w

w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Contents

Quality Control
Reliability and
46.1 Reliability 2
46.2 Quality Control 21
Learning
You will first learn about the importance of the concept of reliability applied to systems and
products. The second Section introduces you to the very important topic of quality control
in production processes. In both cases you will learn how to perform the basic calculations
necessary to use each topic in practice.
outcomes
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Reliability
_
`

46.1
Introduction
Much of the theory of reliability was developed initially for use in the electronics industry where
components often fail with little if any prior warning. In such cases the hazard function or conditional
failure rate function is constant and any functioning component or system is taken to be as new.
There are other cases where the conditional failure rate function is time dependent, often proportional
to the time that the system or component has been in use. The function may be an increasing function
of time (as with random vibrations for example) or decreasing with time (as with concrete whose
strength, up to a point, increases with time). If we can develop a lifetime model, we can use it to
plan such things as maintenance and part replacement schedules, whole system replacements and
reliability testing schedules.
_
`

Prerequisites
Before starting this Section you should . . .
be familiar with the results and concepts met
in the study of probability
understand and be able to use continuous
probability distributions
_

Learning Outcomes
On completion you should be able to . . .
appreciate the importance of lifetime
distributions
complete reliability calculations for simple
systems
state the relationship between the Weibull
distribution and the exponential distribution
2 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
1. Reliability
Lifetime distributions
From an engineering point of view, the ability to predict the lifetime of a whole system or a system
component is very important. Such lifetimes can be predicted using a statistical approach using
appropriate distributions. Common examples of structures whose lifetimes we need to know are
airframes, bridges, oil rigs and at a simpler, less catastrophic level, system components such as cycle
chains, engine timing belts and components of electronic systems such as televisions and computers.
If we can develop a lifetime model, we can use it to plan such things as maintenance and part
replacement schedules, whole system replacements and reliability testing schedules.
We start by looking at the length of use of a system or component prior to failure (the age prior to
failure) and from this develop a denition of reliability. Lifetime distributions are functions of time
and may be expressed as probability density functions and so we may write
P(t) =
_
t
0
p(t) dt
This represents the probability that the system or component fails anywhere between 0 and t.
Key Point 1
The probability that a system or component will fail only after time t may be written as
R(t) = 1 P(t)
The function R(t) is usually called the reliability function.
In practice engineers (and others!) are often interested in the so-called hazard function or conditional
failure rate function which gives the probability that a system or component fails after it has been in
use for a given time. This function may be dened as
H(t) = lim
to
_
P(failure in the interval (t, t + t))/t
P(survival up to time t)
_
=
1
R(t)
lim
to
_
P(t + t) P(t)
t
_
=
1
R(t)
d
dt
P(t)
=
p(t)
R(t)
HELM (2005):
Section 46.1: Reliability
3
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
This gives the conditional failure rate function as
H(t) = lim
to
_
_
_
_
_
t+t
t
p(t) dt/(t)
R(t)
_
_
_
_
=
p(t)
R(t)
Essentially we are describing the rate of failure per unit time for (say) mechanical or electrical
components which have already been in service for a given time. A graph of H(t) often shows a high
initial failure rate followed by a period of relative reliability followed by a period of increasingly high
failure rates as a system or component ages. A typical graph (sometimes called a bathtub graph) is
shown below.
H(t)
0 t
Early life and
random failure
Useful life and random failure
End of life and
random failure
Figure 1
Note that early life and random failure includes failure due to defects being present and that end
of life and random failure includes failure due to ageing.
The reliability of a system or component may be dened as the probability that the system or
component functions for a given time, that is, the probability that it will fail only after the given
time. Put another way, R(t) is the probability that the system or component is still functioning at
time t.
The exponential distribution
We have already met the exponential distribution in the form
f(t) = e
t
, t 0
However, one of the simplest distributions describing failure is the exponential distribution
p(t) =
1

e
t/
, t 0
where, in this case, is the mean time to failure. One property of this distribution is that the hazard
function is a constant independent of time - the good as new syndrome mentioned above. To show
that the probability of failure is independent of age consider the following.
P(t) =
_
t
0
1

e
t/
dt =
1

_
e
t/
_
t
0
= 1 e
t/
4 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Hence the reliability (given that the total area under the curve P(t) = 1 e
t/
is unity) is
R(t) = 1 P(t) = e
t/
Hence, the hazard function or conditional failure rate function H(t) is given by
H(t) =
P(t)
R(t)
=
1

e
t/
e
t/
=
1

which is a constant independent of time.

Another way of looking at this is to consider the probability that failure occurs in the interval (, +t)
given that the system is functioning at time . This probability is
F( + t) F()
R(t)
=
(1 e
(+t)/
) (1 e
/
e
/
= 1 e
t/
This is just the probability that failure occurs in the interval (0, t) and implies that ageing has no
eect on failure. This is sometimes referred to as the good as new syndrome.
It is worth noting that in the modelling of many complex systems it is assumed that only random
component failures are important. This enables us to assume the use of the exponential distribution
since initial failures are removed by a running-in process and the time to ultimate failure is usually
long.
Example 1
The lifetime of a modern low-wattage electronic light bulb is known to be expo-
nentially distributed with a mean of 8000 hours.
Find the proportion of bulbs that may be expected to fail before 7000 hours use.
Solution
We know that = 8000 and so
P(t) =
_
t
0
1
8000
e
t/8000
dt
=
1
8000
_
8000e
t/8000
_
t
0
= 1 e
t/8000
Hence P(7000) = 1 e
7000/8000
= 1 e
0.675
= 0.4908 and we expect that about 49% of the
bulbs will fail before 7000 hours of use.
HELM (2005):
Section 46.1: Reliability
5
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Task
A particular electronic device will only function correctly if two essential compo-
nents both function correctly. The lifetime of the rst component is known to be
exponentially distributed with a mean of 6000 hours and the lifetime of the second
component is known to be exponentially distributed with a mean of 7000 hours.
Find the proportion of devices that may be expected to fail before 8000 hours use.
State clearly any assumptions you make.
Your solution
Answer
The assumption made is that the components operate independently.
For the rst component P(t) = 1e
t/6000
so that P(8000) = 1e
8000/6000
= 1e
4/3
= 0.7364
For the second component P(t) = 1 e
t/7000
so that P(8000) = 1 e
8000/7000
= 0.6811
The probability that the device will continue to function after 8000 hours use is given by an expression
of the form P(A B) = P(A) + P(B) P(A B)
Hence the probability that the device will continue to function after 8000 hour use is
0.7364 + 0.6811 0.7364 0.6811 = 0.916
and we expect just under 92% of the devices to fail before 8000 hours use.
An alternative answer may be obtained more directly by using the reliability function R(t):
Answer
The assumption made is that the components operate independently.
For the rst component R(t) = e
t/6000
so that R(8000) = e
8000/6000
= e
4/3
= 0.2636
For the second component R(t) = e
t/7000
so that R(8000) = e
8000/7000
= 0.3189
The probability that the device will continue to function after 8000 hour use is given by
0.2636 0.3189 = 0.0841
Hence the probability that the device will fail before 8000 hours use is 1 0.0841 = 0.916
and we expect just under 92% of the devices to fail before 8000 hours use.
6 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
2. System reliability
It is reasonable to ask whether, in designing a system, an engineer should design a system using
components in series or in parallel. The engineer may not have a choice of course! We may represent
a system consisting of n components say C
1
, C
2
. . . , C
n
with reliabilities (these are just probability
values) R
1
, R
2
. . . , R
n
respectively as series and parallel systems as shown below.
C
1
C
1
C
2 C
n
C
2
C
n
Series Design Parallel Design
Figure 2
With a series design, the system will fail if any component fails. With a parallel design, the system
will work as long as any component works.
Assuming that the components are independent, we can express the reliability of the series design as
R
Series
= R
1
R
2
R
n
simply by multiplying the probabilities.
Since each reliability value is less than one, we may conclude that a series design is less reliable than
its least reliable component.
Similarly (although by no means as clearly!), we can express the reliability of the parallel design as
R
Parallel
= 1 (1 R
1
)(1 R
2
) . . . (1 R
n
)
The derivation of this result is illustrated in Example 3 below for the case n = 3 . In this case,
the algebra involved in fairly straightforward. We can conclude that the parallel design is at least as
reliable as the most reliable component.
Engineers will sometimes include redundantcomponents in parallel to improve reliability. The spare
wheel of a car is a well known example.
HELM (2005):
Section 46.1: Reliability
7
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Example 2
Series design
Consider the three components C
1
, C
2
and C
3
with reliabilities R
1
, R
2
and R
3
connected in series as shown below
C
1
C
2 C

Find the reliability of the system where R

1
= 0.2, R
2
= 0.3 and R
3
= 0.4.
Solution
Since the components are assumed to act independently, we may clearly write
R
Series
= R
1
R
2
R
3
Taking R
1
= 0.2, R
2
= 0.3 and R
3
= 0.4 we obtain the value R
Series
= 0.2 0.3 0.4 = 0.024
Example 3
Parallel design
Consider the three components C
1
, C
2
and C
3
with reliabilities R
1
, R
2
and R
3
connected in parallel as shown below
C
1
C
2
C

Find the reliability of the system where R

1
= 0.2, R
2
= 0.3 and R
3
= 0.4.
Solution
Observing that F
i
= 1R
i
, where F
i
represents the failure of the ith component and R
i
represents
the reliability of the ith component we may write
F
System
= F
1
F
2
F
3
R
System
= 1 F
1
F
2
F
3
= 1 (1 R
1
)(1 R
2
)(1 R
3
)
Again taking R
1
= 0.2, R
2
= 0.3 and R
3
= 0.4 we obtain
F
System
= 1 (1 0.2)(1 0.3)(1 0.4) = 1 0.336 = 0.664
8 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Hence series system reliability is less than any of the component reliabilities and parallel system
reliability is greater than any of the component reliabilities.
Task
Consider the two components C
1
and C
2
with reliabilities R
1
and R
2
connected
in series and in parallel as shown below. Assume that R
1
= 0.3 and R
2
= 0.4.
C
1
C
2
C
1
C
2
Series conguration Parallel conguration
Let R
Series
be the reliability of the series conguration and R
Parallel
be the reliability
of the parallel conguration
(a) Why would you expect that R
Series
< 0.3 and R
Parallel
> 0.4?
(b) Calculate R
Series
(c) Calculate R
Parallel
Your solution
Answer
(a) You would expect R
Series
< 0.3 and R
Parallel
> 0.4 because R
Series
is less than any of the
component reliabilities and R
Parallel
is greater than any of the component reliabilities.
(b) R
Series
= R
1
R
2
= 0.3 0.4 = 0.12
(c) R
Parallel
= R
1
R
2
R
1
R
2
= 0.3 + 0.4 0.3 0.4 = 0.58
HELM (2005):
Section 46.1: Reliability
9
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
3. The Weibull distribution
The Weibull distribution was rst used to describe the behaviour of light bulbs as they undergo the
ageing process. Other applications are widespread and include the description of structural failure,
ball-bearing failure and the failure of a variety of electronic components.
Key Point 2
If the random variable X is dened by a probability density function of the form
f(x) = (x)
1
e
(x)

then X is said to follow a Weibull distribution.

The hazard function or conditional failure rate function H(t), which gives the probability that a
system or component fails after it has been in use for a given time, is constant for the exponential
distribution but for a Weibull distribution is proportional to x
1
. This implies that = 1 gives a
constant hazard, < 1 gives a reducing hazard and > 1 gives an increasing hazard. Note that
is simply a scale factor while the case = 1 reduces the Weibull distribution to
f(x) = e
x
which you may recognize as one form of the exponential distribution.
Figure 3 below shows the Weibull distribution for various values of . For simplicity, the graphs
assume that = 1. Essentially the plots are of the function f(x) = x
1
e
x

= 1
= 2
= 4
= 8
1
2
3
0
1 2 3 4
4
The Weibull distribution with = 1, 2, 4 and 8
Figure 3
In applications of the Weibull distribution, it is often useful to know the cumulative distribution
function (cdf). Although it is not derived directly here, the cdf F(x) of the Weibull distribution,
whose probability density function is f(x) = (x)
1
e
(x)

, is given by the function

10 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
F(x) = 1 e
(x)

The relationship between f(x) and F(x) may be seen by remembering that
f(x) =
d
dx
F(x)
Dierentiating the cdf F(x) = 1 e
(x)

gives the result

f(x) =
d
dx
F(x) =
(x)

e
(x)

x
=

x
1
e
(x)

= (x)
1
e
(x)

Mean and variance of the Weibull distribution

The mean and variance of the Weibull distribution are quite complicated expressions and involve the
use of the Gamma function, (x) . The outline explanations given below denes the Gamma function
and shows that basically in terms of the application here, the function can be expressed as a simple
factorial.
It can be shown that if the random variable X follows a Weibull distribution dened by the probability
density function
f(x) = (x)
1
e
(x)

then the mean and variance of the distribution are given by the expressions
E(X) =
1

(1 +
1

) and V(X) =
_
1

_
2
_
(1 +
2

)
_
(1 +
1

)
_
2
_
where (r) is the gamma function dened by
(r) =
_

0
x
r1
e
x
dx
Using integration by parts gives a fairly straightforward identity satised by the gamma function is:
(r) = (r 1)(r 1)
Using this relationship tells us that
(r) = (r 1)(r 2)(r 3) . . . (1) for integral r
but
(1) =
_

0
x
11
e
x
dx =
_

0
e
x
dx =
_
e
x
_

0
= 1
so that
(r) = (r 1)(r 2)(r 3) . . . (3)(2)(1) = (r 1)! for integral r
Essentially, this means that when we calculate the mean and variance of the Weibull distribution
using the expressions
E(X) =
1

(1 +
1

) and V(X) =
_
1

_
_
(1 +
2

)
_
(1 +
1

)
_
2
_
we are evaluating factorials and the calculation is not as dicult as you might imagine at rst sight.
Note that to apply the work as presented here, the expressions
HELM (2005):
Section 46.1: Reliability
11
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
(1 +
1

) and (1 +
2

)
must both take positive integral values (i.e.
1

must be an integer) in order for the factorial calculation

to make sense. If the above expressions do not assume positive integral values, you will need to know
much more about the values and behaviour of the Gamma function in order to do calculations. In
practice, we would use a computer of course. As you might expect, the Gamma function is tabulated
to assist in such calculations but in the examples and exercises set in this Workbook, the values of
1

will always be integral.

Example 4
The main drive shaft of a pumping engine runs in two bearings whose lifetime
follows a Weibull distribution with random variable X and parameters = 0.0002
and = 0.5.
(a) Find the expected time that a single bearing runs before failure.
(b) Find the probability that any one bearing lasts at least 8000 hours.
(c) Find the probability that both bearings last at least 6000 hours.
Solution
(a) We know that E(X) =
1

(1 +
1

) = 5000 (1 + 2) = 5000 2 = 10000 hours.

(b) We require P(X > 8000) , this is given by the calculation:
P(X > 8000) = 1 P(8000) = 1 (1 e
(0.00028000)
0.5
)
= e
(0.00028000)
0.5
= e
1.265
= 0.282
(c) Assuming that the bearings wear independently, the probability that both bearings last
at least 6000 hours is given by {P(X > 6000)}
2
. But P(X > 6000) is given by the
calculation
P(X > 6000) = 1 P(6000) = 1 (1 e
(0.00026000)
0.5
) = 0.335
so that the probability that both bearings last at least 6000 hours is given by
{P(X > 6000)}
2
= 0.335
2
= 0.112
12 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Task
A shaft runs in four roller bearings the lifetime of each of which follows a Weibull
distribution with parameters = 0.0001 and = 1/3.
(a) Find the mean life of a bearing.
(b) Find the variance of the life of a bearing.
(c) Find the probability that all four bearings last at least 50,000 hours.
State clearly any assumptions you make when determining this proba-
bility.
Your solution
Answer
(a) We know that E(X) =
1

(1 +
1

) = 10000 (1 + 3) = 60000 2 = 10000 hours.

(b) We know that V(X) =
_
1

_
_
(1 +
2

)
_
(1 +
1

_
2
_
so that
V(X) = (10000)
2
{(1 + 6) ((1 + 3)]
2
}
= (10000)
2
{(7) ((4))
2
}
= (10000)
2
{6! (3!)
2
} = 684(10000)
2
= 6.84 10
10
(c) The probability that one bearing will last at least 50,000 hours is given by the calculation
P(X > 50000) = 1 P(50000)
= 1 (1 e
(0.000150000)
0.5
)
= e
5
0.5
= 0.107
Assuming that the four bearings have independent distributions, the probability that all
four bearings last at least 50,000 hours is (0.107)
4
= 0.0001
HELM (2005):
Section 46.1: Reliability
13
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Exercises
1. The lifetimes in hours of certain machines have Weibull distributions with probability density
function
f(t) =
_
0 (t < 0)
(t)
1
exp{(t)

} (t 0)
(a) Verify that
_
t
0
(u)
1
exp{(u)

}du = 1 exp{(t)

}.
(b) Write down the distribution function of the lifetime.
(c) Find the probability that a particular machine is still working after 500 hours of use if
= 0.001 and = 2.
(d) In a factory, n of these machines are installed and started together. Assuming that
failures in the machines are independent, nd the probability that all of the machines are
still working after t hours and hence nd the probability density function of the time till
the rst failure among the machines.
2. If the lifetime distribution of a machine has hazard function h(t), then we can nd the reliability
function R(t) as follows. First we nd the cumulative hazard function H(t) using
H(t) =
_
t
0
h(t) dt then the reliability is R(t) = e
H(t)
.
(a) Starting with R(t) = exp{H(t)}, work back to the hazard function and hence conrm
the method of nding the reliability from the hazard.
(b) A lifetime distribution has hazard function h(t) =
0
+
1
t +
2
t
2
. Find
(i) the reliability function.
(ii) the probability density function.
(c) A lifetime distribution has hazard function h(t) =
1
t + 1
. Find
(i) the reliability function;
(ii) the probability density function;
(iii) the median.
What happens if you try to nd the mean?
(d) A lifetime distribution has hazard function h(t) =

t + 1
, where > 0 and > 1.
Find
(i) the reliability function.
(ii) the probability density function.
(iii) the median.
(iv) the mean.
14 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Exercises continued
3. A machine has n components, the lifetimes of which are independent. However the whole
machine will fail if any component fails. The hazard functions for the components are
h
1
(t), . . . , h
n
(t). Show that the hazard function for the machine is
n

i=1
h
i
(t).
4. Suppose that the lifetime distributions of the components in Exercise 3 are Weibull distributions
with scale parameters
1
, . . . ,
n
and a common index (i.e. shape parameter) so that the
hazard function for component i is
i
(
i
t)
1
. Find the lifetime distribution for the machine.
5. (Dicult). Show that, if T is a Weibull random variable with hazard function (t)
1
,
(a) the median is M(T) =
1
(ln 2)
1/
,
(b) the mean is E(T) =
1
(1 +
1
)
(c) the variance is V(T) =
2
{(1 + 2
1
) [(1 +
1
)]
2
}.
Note that (r) =
_

0
x
r1
e
x
dx. In the integrations it is helpful to use the substitution u = (t)

.
HELM (2005):
Section 46.1: Reliability
15
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Answers
1.
(a)
d
du
(exp{(u)

})
so
_
t
0
(u)
1
exp{(u)

}du =
_
exp{(u)

}
_
t
0
= 1 exp{(t)

}.
(b) The distribution function is F(t) = 1 exp{(t)

}.
(c) R(500) = exp{(0.001 500)
2
} = 0.7788
(d) The probability that all of the machines are still working after t hours is
R
n
(t) = {R(t)}
n
= exp{(t)
n
}.
Hence the time to the rst failure has a Weibull distribution with replaced by n. The
pdf is
f
n
(t) =
_
0 (t < 0)
n(t)
n1
exp{(t)
n
} (t 0)
2.
(a) The distribution function is
F(t) = 1 R(t) = 1 e
H(t)
so the pdf is f(t) =
d
dt
F(t) = h(t)e
H(t)
.
Hence the hazard function is
h(t) =
f(t)
R(t)
=
h(t)e
H(t)
e
H(t)
= h(t).
(b) A lifetime distribution has hazard function h(t) =
0
+
1
t +
2
t
2
.
(i) Reliability R(t).
H(t) =
_

0
t +
1
t
2
/2 +
2
t
3
/3
_
t
0
=
0
t +
1
t
2
/2 +
2
t
3
/3
R(t) = exp{(
0
t +
1
t
2
/2 +
2
t
3
/3)}.
(ii) Probability density function f(t).
F(t) = 1 R(t) f(t) = (
0
+
1
t +
2
t
2
) exp{(
0
t +
1
t
2
/2 +
2
t
3
/3)}.
16 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Answers
(c) A lifetime distribution has hazard function h(t) =
1
t + 1
.
(i) Reliability function R(t).
H(t) =
_
t
0
1
u + 1
du =
_
ln(u + 1)
_
t
0
= ln(t + 1)
R(t) = exp{H(t)} =
1
t + 1
(ii) Probability density function f(t). F(t) = 1 R(t), f(t) =
1
(t + 1)
2
(iii) Median M.
1
M + 1
=
1
2
so M = 1.
To nd the mean: E(T + 1) =
_

0
1
t + 1
dt which does not converge, so neither does E(T).
(d) A lifetime distribution has hazard function
h(t) =

t + 1
, where > 0 and > 1.
(i) Reliability function R(t).
H(t) =
_
t
0

t + 1
dt =
_
ln(t + 1)
_
t
0
= ln(t + 1)
R(t) = exp{ ln(t + 1)} = (t + 1)

(ii) Probability density function f(t).

F(t) = 1 R(t)
f(t) = (t + 1)
(1+)
(iii) Median M.
(M + 1)

= 2
1
M + 1 = 2
1/
M =
1
(2
1/
1)
(iv) Mean.
E(T + 1) =
_

0
(t + 1)

dt
=

1
_
(t + 1)
1
_

0
=

1
E(T) =
1
_

1
1
_
=
1
( 1)
1
HELM (2005):
Section 46.1: Reliability
17
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Answers
3. For each component
H
i
(t) =
_
t
0
h
i
(t) dt
R
i
(t) = exp{H
i
(t)}
For machine
R(t) =
n

i=1
R
i
(t) = exp{
n

i=1
H
i
(t)}
the probability that all components are still working.
F(t) = 1 R(t)
f(t) = exp{
n

i=1
H
i
(t)}
n

i=1
H
i
(t)
h(t) = f(t)/R(t) =
n

i=1
h
i
(t).
4. For each component
h
i
(t) =
i
(
i
t)
1
H
i
(t) =
_
t
0

i
(
i
t)
1
dt = (
i
t)

Hence, for the machine,

n

i=1
H
i
(t) = t

i=1

i
R(t) = exp
_
t

i=1

i
_
F(t) = 1 R(t)
f(t) = t
1
n

i=1

i
exp
_
t

i=1

i
_
h(t) = (t)
1
so we have a Weibull distribution with index and scale parameter such that

=
n

i=1

i
.
18 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Answers
5. Suppose the distribution function is
F(t) = 1 e
(t)

.
Then the pdf is
f(t) =
d
dt
F(t) = (t)
1
e
(t)

and the hazard function is

H(t) =
f(t)
R(t)
=
f(t)
1 F(t)
= (t)
1
as required.
(a) At the median, M, F(M) = 0.5. So
1 e
(M)

= 0.5
e
(M)

= 0.5
(M)

= ln(0.5) = ln(2)
M = (ln 2)
1/
M =
1
(ln 2)
1/
(b) The mean is
E(T) =
_

0
t f(t) dt =
_

0
t (t)
1
e
(t)

dt
=
_

0
(t)

e
(t)

dt
Let u = (t)

, so du/dt = (t)
1
and du = (t)
1
dt. Then
E(T) =
_

0

1
u
1/
e
u
du
=
1
_

0
u
1+1/1
e
u
du
=
1
(1 + 1/)
HELM (2005):
Section 46.1: Reliability
19
www.ebookcraze.blogspot.com
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
w
w
w
.
e
b
o
o
k
c
r
a
z
e
.
b
l
o
g
s
p
o
t
.
c
o
m
Answers
5. continued
(c) To nd the variance we rst nd
E(T
2
) =
_

0
t
2
f(t) dt =
_

0
t
2
(t)
1
e
(t)

dt
=
_

0

1
(t)
+1
e
(t)

dt
Let u = (t)

, so du/dt = (t)
1
and du = (t)
1
dt. Then
E(T
2
) =
_

0

2
u
2/
e
u
du
=
2
_

0
u
1+2/1
e
u
du
=
2
(1 + 2/)
So
V(T) = E(T
2
) {E(T)}
2
=
2
(1 + 2/)
2
{(1 + 1/)}
2
=
2
{(1 + 2/) {(1 + 1/)}
2
}
20 HELM (2005):
Workbook 46: Reliability and Quality Control
www.ebookcraze.blogspot.com